The invention relates to the field of stereoscopic vision devices.
Stereoscopic vision is required and used in a multiplicity of applications, including but not limited to determining three-dimensional (3D) information in an environment for automatically navigating a robot therein. The task is particularly complex in an indoor or another environment in which the robot is surrounded by multiple static or dynamic objects at various and changing distances from the robot.
In order to obtain three-dimensional information of an environment, some systems use a pair of cameras whose fields of view (FOVs) have part or full overlap. Depth information may be obtained by registering the two images, comprising identifying one or more objects or locations appearing in the two images, and deducing depth information from their respective locations, size, orientation or other parameters.
The foregoing examples of the related art and limitations related therewith are intended to be illustrative and not exclusive. Other limitations of the related art may become apparent to those of skill in the art upon a reading of the specification and a study of the figures.